9

Open-source hardware for nature conservation

FEATURE

Wilderness

Project

n 2015, Shah Selbe joined fellow National

Geographic explorer Dr Steve Boyes on an

expedition to survey the vast wilderness

surrounding the Angolan Highlands to

collect environmental data and trace the

water that feeds the Okavango Delta from

its source.

At over 720 000 square kilometres, the greater

Okavango River Basin is the largest freshwater

wetland in southern Africa and the primary source of

water for over a million people. Its delta, located in

northern Botswana, is one of Africa’s wealthiest

places for biodiversity, and home to the world’s

largest remaining elephant population, as well as

populations of some of the world’s most endangered

animals, such as cheetah, white rhino, black rhino,

African wild dog, and lion.

The Kubango and the Cuito rivers supply nearly all

of the water that flows into the Okavango Delta,

meaning the protection of these two rivers, which

originate in southern Angola, is vital to preserving

the richness of wildlife in this untamed region.

During the expedition, the research team were no

longer required to use pH strips or manually check

sensor readings, as Selbe had built a wireless sensor

network to automate the process. “Shah took us

from little strips and pieces of paper – writing down

the water quality as we go down – to environmental

sensor platforms,” explained Dr Boyes.

In the past, any data that was collected needed to

be taken back to the lab and analysed before it could

be acted upon. With the Okavango Wilderness

Project, much of that data is now being live

streamed, allowing for a significantly shorter

turnaround between discovering problems and

recommending solutions. The team has also been

working to collect higher resolution data, including

360-degree videos and wildlife photographs with

references to the time and GPS location they were

taken. When combined with data from traditional

sensor technology (water and air temperature,

humidity, water flow rate, etc.), a better overall

understanding of the ecosystem and its status can

be achieved.

The wireless sensor network itself consists of a

Raspberry Pi running an open-source Python script

to process the data generated from multiple remote

Arduino nodes. The Raspberry Pi acts as a WiFi

gateway, and directly uploads data to the web server

using JSON. In some particularly remote locations,

the remote nodes can send data using the Twilio API

over a cellular network.

I

Above

The Okavango Delta

on the evening this

post was uploaded

Image Credits

Shah Selbe CC BY-SA